When data is lost in translation
Sponsor Content From
Matt Cardwell, PhD
Chief Product Officer, Intelligent Medical Objects
Often times, medical messages get lost in translation despite the best efforts of clinicians who document patient care. This is a problem not only for individual Electronic Health Record (EHR) users, but also for systems that increasingly depend on interoperability in order to send medical information between providers.
Behind both of these problems lies one key culprit – that providing medical care requires many languages. From the natural-speech style of the clinician’s diagnosis to the rigid computer-speak of coding systems, such as ICD-10-CM or SNOMED®, there are simply too many ways to say the same thing. When systems do not have an effective translator at this medical Tower of Babel, language and nuance are lost, creating headaches for both providers and physicians.
For the most part, it’s easy to see how this can negatively impact individual EHR users during the course of their workdays. Indeed, much has been written on the topic. However, even if data is captured with full clinical intent — meaning physicians are able to document exactly what they want to say — it is possible for that intent to be lost when transmitting data to other information systems (like another EHR).
Two systems at work
Even if they are not “fluent”, most EHR users have a base familiarity with or awareness of the many “languages”, like CPT® and ICD-10-CM codes, that are used within their system. However, transmitting data to any system outside of the initial EHR requires yet another translation – and another opportunity for data fidelity to be degraded.
On a global scale, this translation standard starts with Health Level-7 (HL-7), an independent international organization that creates suggested modes of transmission for data traveling between health IT systems. Governments then frequently adapt these standards to create their own, country-specific guidelines.
These messaging standards, called C-CDA or FHIR in the United States, can contribute to degradation of data’s semantic content if they are implemented improperly or without enough attention to detail. This often occurs when systems are forced to reduce complex health data into a single, primary code. The recipients of the data then end up working with inaccurate or incomplete information even though the clinical message was clearly articulated during the initial documentation.
The ‘heart’ of the issue
It’s clear that this lack of accurate communication can have major implications for any organization wishing to successfully transition medical information between providers. Just as important, though, is the potential impact on patient care, highlighted in the following example. When a patient has a diagnosis of “breast cancer metastasized to pelvis,” the diagnosis must be reduced to only a single SNOMED code to comply with the C-CDA messaging standard. As a result, only the primary breast cancer diagnosis is communicated when the patient’s chart is sent from one system to another. Any analytics performed using the information in the downstream system would therefore lose the staging (metastatic) component of the original diagnosis. Moreover, any application trying to assess this patient’s clinical situation at a future point in her care would likewise be compromised.
Understanding how these translation systems affect “the” patient’s care, not patient care in the abstract, can help us to understand the consequences of not having an effective translation technology in place. It’s critical that we recognize the importance of maintaining the integrity of our data, whether it’s within a given system or in the process of transmission from one system to another. As we look to the next era of EHR improvement, remembering how this impacts not just healthcare professionals, but also the patients they serve, can help shine the light on better solutions in the future.
SNOMED and SNOMED CT are registered trademarks of SNOMED International
CPT is a registered trademark of the American Medical Association. All rights reserved.